Please cite this article in press as: R. Surez, et al.

, Intangible cultural heritage: The sound of the Romanesque cathedral of Santiago de

Compostela, Journal of Cultural Heritage (2014), http://dx.doi.org/10.1016/j.culher.2014.05.008
ARTICLE IN PRESS
G Model
CULHER-2864; No. of Pages 5
Journal of Cultural Heritage xxx (2014) xxxxxx
Available online at
ScienceDirect
www.sciencedirect.com
Case study
Intangible cultural heritage: The sound of the Romanesque cathedral
of Santiago de Compostela
Rafael Surez

, Alicia Alonso , Juan J. Sendra

Instituto Universitario de Arquitectura y Ciencias de la Construccin, Escuela Tcnica Superior de Arquitectura, Universidad de Sevilla, Av. Reina Mercedes
2, Seville 41012, Spain
a r t i c l e i n f o
Article history:
Received 4 November 2013
Accepted 15 May 2014
Available online xxx
Keywords:
Worship acoustics
Cultural heritage
Religious music
Romanesque cathedral
Cathedral of Santiago de Compostela
a b s t r a c t
Since the end of the twentieth century the increasing importance of cultural heritage has been comple-
mented with the recognition and protection of intangible heritage. This new approach to cultural heritage
requires scientic knowledge supported by new technologies. This context is used as the starting point for
furthering knowledge of the relationship between architecture, liturgy and music in Romanesque church
architecture the rst artistic style to become widespread in the West by including a new intangible
dimension: sound. This case study of the emblematic cathedral of Santiago de Compostela uses computer
simulation to expose the acoustic behaviour of the original Romanesque space now covered by Baroque
elements and its effect on the functional and spatial structure.
2014 Elsevier Masson SAS. All rights reserved.
1. Research aim
The main aim of this work is to analyze the relationship
between architecture, music and liturgy through the acoustics of
the Romanesque cathedral of Santiago de Compostela, which is
usedfor bothreligious worshipandfor the preservationof religious
relics. This unique cathedral shows no physical traces of the medi-
aeval Roman liturgy or the Romanesque space, which have been
concealed by Baroque interventions. However, it is possible to cre-
ate an acoustic model of the original Romanesque space and use
computer simulation to establish the main parameters that deter-
mine the acoustic behavior of this space. It is even possible to use
auralizations to virtually recreate the sound of the Romanesque
cathedral using acoustic archaeology [1]. This technology aims to
recover one of the lost or forgotten intangible cultural heritage val-
ues of the cathedral of Santiago, that of sound, and help establish
its status as a factor for cultural dynamization.
2. Introduction
The late eleventhcentury liturgical reformationandthe increas-
ing importance of religious orders under the hegemony of Cluny
set the stage for the birth of Romanesque, the rst artistic style
to be widespread in the West. This new interpretation of space

Corresponding author. Tel.: +34 954 55 95 17; fax: +34 954 55 70 18.
E-mail address: rsuarez@us.es (R. Surez).
encouraged the extraordinary development of Christian music,
both that of Gregorian chant and the birth of polyphony.
The study of Romanesque religious architecture should move
away fromtraditional studies to concentrate on its physical reality,
its visual and style variations, interpreting the building as a whole,
thus opening upto newhistorical and heritage readings, to changes
in its functional history [2] and to the incorporation of the liturgy
[3] and music [4] as intangible values to be protected [5].
The organization and purpose of the liturgy and the devel-
opment of music determined the conguration and evolution of
Romanesque architecture. This close relationship between archi-
tecture, music and liturgy demanded a new approach that focuses
on the multi-sensory experience of architecture, complementing
thevisual spacewiththesoundeld, andincorporatingtheacoustic
dimension through the reconstruction of sound.
Specialist literature includes researchwithdifferent approaches
to this relationship, either through the analysis of the relationship
between architecture and acoustics in religious buildings in their
current condition [6,7] fromthe perspective of the acoustic evolu-
tion connecting liturgy, music and architecture [810], or through
the acoustic study of these interconnections in the origins of Chris-
tianity with layouts which have clearly undergone transformation
or may no longer exist [11].
Thecathedral of SantiagodeCompostelais anexceptional exam-
ple as it has been a universal pilgrimage destination for Christians
since the Middle Ages. The cathedral, along with the old town, was
declared World Heritage Site by UNESCO in 1985, and the Way to
Santiago, also declared a World Heritage Site in 1993, formed part
http://dx.doi.org/10.1016/j.culher.2014.05.008
1296-2074/ 2014 Elsevier Masson SAS. All rights reserved.
Please cite this article in press as: R. Surez, et al., Intangible cultural heritage: The sound of the Romanesque cathedral of Santiago de
Compostela, Journal of Cultural Heritage (2014), http://dx.doi.org/10.1016/j.culher.2014.05.008
ARTICLE IN PRESS
G Model
CULHER-2864; No. of Pages 5
2 R. Surez et al. / Journal of Cultural Heritage xxx (2014) xxxxxx
of the most important mediaeval religious and cultural movement,
the most efcient channel for the construction of Europe.
3. Methods
The latest techniques for acoustic prediction were applied once
the acoustic models of the Romanesque cathedral had been con-
structedreproducing its geometryandthe distributionof the sound
absorption-dispersion of the materials had been established, and
the receivers positioned according to occupation, sound source
position and the different liturgical uses of each space.
The acoustic model simulations provided the room impulse
responses (RIR) for each receiver, implementing high quality aural-
izations togenerateall theinformationontheacoustic conditions of
the room. Once RIRs were obtained it was possible to calculate the
most important acoustic parameters of the Romanesque cathedral
and generate audio signal samples that could be used for subjective
valorisations or auralizations. This procedure is used for acoustic
archaeology, a virtual acoustic reconstructionof these spaces from
the past, incorporating acoustic characteristics as animportant part
of their intangible heritage. The software used for the simulation
was CATT-Acoustic, version 9.0, with TUCT (The Universal Cone
Tracer) calculation engine version 1.0 [12].
This methodology aims to assess the differences in the acous-
tic behaviour of each spatial conguration according to purpose,
instead of merely determining the acoustic quality of this space
according to current roomacoustic evaluation criteria.
3.1. Case study
The main documentary source for updating the image of the
cathedral of Santiago to bring to life the festive, liturgical and musi-
cal representation at the time of consecration (1211) is the Codex
Calixtinus [13]. This codex includes the liturgy and music produced
to honor the apostle Santiago on major feasts. The Romanesque
space is documented in research by K.J. Conant [14], and other
research projects currently being carried out by Wunderwald and
Rffer [15] and Dagenais [16].
The Romanesque cathedral of Santiago de Compostela follows
the model of pilgrimage churches of French origin, both in terms
of style and the majority of its functions, and is one of the most
beautiful typically Romanesque mediaeval churches. It was built
mostly out of granite, with a Latin cross oorplan, and three side
naves whose measurements are included in the Codex (Fig. 1). The
main nave is covered by barrel vaults while the side naves are cov-
ered by ribbed vaults. The naves are separated by columns, with a
tribune over the side naves covered by a quarter round barrel vault.
The formal plan is based on monastic buildings, fromwhich its
use is adapted, with a shallow presbytery for the main altar and
a wide choir in the central nave for the chanting of divine ofce.
This stone choir, executed by Master architect Mateo, follows the
model of Cluny andwas to become a permanent element inSpanish
cathedrals. It is located in the rst three stretches of the nave from
the transept, with a fourth stretch assigned to the leedoiro, a high
tribune used for sermons in ceremonies.
3.2. The acoustic model
The acoustic models used for the simulations with CATT-
Acoustic had 5600 planes and a volume of 48,202m
3
. These
simulations make it possible to ascertain how the sound eld is
modied by the mass presence of a congregation on major feasts.
Three ominidirectional sources were placed at a height of 1.5m
above ground level, positioned according to the main liturgical
functions of the cathedral: the presbytery for the celebration of the
mass at the altar (S1), the choir for chanting by the canons (S2) and
Fig. 1. (a) Longitudinal section and (b) oorplan of the Romanesque cathedral of
Santiago de Compostela. Zoning and distribution of sound sources and receivers; (c)
acoustic model.
the leedoiro for speech (S3). Twenty-one receiver positions were
adopted(Fig. 1, Table1) distributedthroughout thetransept, ambu-
latory, choir, trascoro, sidenaves andtribune, all at 1.60meters from
the ground, a height which has been set assuming that the congre-
gation was standing. Each surface of the building was assigned a
material with its corresponding absorption and scattering coef-
cient (Table 2), also incorporating the absorption of sound energy
by the volume of air in this large space.
The results of the main acoustic parameters and the subjective
perception dened in ISO 3382-1:2009 [17] were assessed follow-
ing the simulation of the acoustic models generated. The physical
Table 1
Receiver location.
Zones Receivers
Transept R1, R2, R3, R4, R5, R6
Ambulatory R7
Trascoro R8, R9, R10, R11
Aisle R12, R13, R14, R15
Choir R16, R17
Tribune (rst oor)
a
R18, R19, R20, R21
a
Placed above points 12 to 15 respectively.
Please cite this article in press as: R. Surez, et al., Intangible cultural heritage: The sound of the Romanesque cathedral of Santiago de
Compostela, Journal of Cultural Heritage (2014), http://dx.doi.org/10.1016/j.culher.2014.05.008
ARTICLE IN PRESS
G Model
CULHER-2864; No. of Pages 5
R. Surez et al. / Journal of Cultural Heritage xxx (2014) xxxxxx 3
Table 2
Absorption coefcients for materials used.
Material Surface (%) 125Hz 250Hz 500Hz 1kHz 2kHz 4kHz
Granite (oor
a
, walls
a
, choir
b
) 61.1 0.02 0.02 0.03 0.04 0.05 0.05
Plaster vaults
a
20.4 0.02 0.02 0.03 0.03 0.04 0.05
Windows
a
1.3 0.13 0.12 0.08 0.07 0.06 0.04
Wood
a
3 0.09 0.09 0.08 0.08 0.10 0.07
Audience
b
area, 1 person/m
2
2.5 0.16 0.29 0.55 0.80 0.92 0.90
Audience
b
area, 0.72 person/m
2
11.7 0.10 0.20 0.41 0.65 0.75 0.71
a
Scattering coefcients 0.12, 0.13, 0.14, 0.15, 0.16, 0.17.
b
Scattering coefcients 0.55, 0.60, 0.65, 0.70, 0.75, 0.80.
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
Empty
Ocuppied S1
125 250 500 1000 2000 4000
T
3
0
(
s
)
Frequency (Hz)
Fig. 2. Octave-band values of average reverberation time T
30
: empty and occupied.
reverberation of the space was calculated through reverberation
time (T
30
). The perceived reverberation was correlated with Early
Decay Time (EDT). The clarity perceived for music was related with
clarity(C
80
). The perceivedsoundlevel was assessedthroughsound
strength (G) that measures the amplication capacity of the build-
ing, making it possible to assess the uniformity of the sound eld.
Finally, speech clarity was assessed using the denition parame-
ter (D
50
). All parameters were spectrally averaged following ISO
3382-1:2009.
4. Results
The indoor acoustic qualities of the Romanesque cathedral, with
its enormous volume of 48202 m
3
and highly reectant materials,
encouraged high reverberation times with mid frequency values,
T
m
, of 8.15 s. The mass occupation of the cathedral in the transept
2.5
2.7
2.9
3.1
3.3
3.5
3.7
3.9
1.00 2.00 3.00 4.00
Transept
Ambulatory
Trascoro
Aisle
Choir
Tribune
T
3
0
(
s
)
S1
EDT (s)
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
125 250 500 1000 2000 4000
Transept
Ambulatory
Trascoro
Aisle
Choir
Tribune
Frequency (Hz)
T
3
0
(
s
)
S1
2.5
2.7
2.9
3.1
3.3
3.5
3.7
3.9
1.00 2.00 3.00 4.00
T
3
0
(
s
)
S2
EDT (s)
0.0
2.0
4.0
6.0
8.0
10.0
12.0
125 250 500 1000 2000 4000
Frequency (Hz)
T
3
0
(
s
)
S2
2.5
2.7
2.9
3.1
3.3
3.5
3.7
3.9
1.00 2.00 3.00 4.00 5.00
EDT (s)
T
3
0
(
s
)
S3
(a)
0.0
2.0
4.0
6.0
8.0
10.0
12.0
125 250 500 1000 2000 4000
Frecuency (Hz)
T
3
0
(
s
)
S3
(b)
Fig. 3. (a) T
30
/EDT by source and receiver. (b) Octave-band values of average reverberation time T
30
by zones and source.
Please cite this article in press as: R. Surez, et al., Intangible cultural heritage: The sound of the Romanesque cathedral of Santiago de
Compostela, Journal of Cultural Heritage (2014), http://dx.doi.org/10.1016/j.culher.2014.05.008
ARTICLE IN PRESS
G Model
CULHER-2864; No. of Pages 5
4 R. Surez et al. / Journal of Cultural Heritage xxx (2014) xxxxxx
-8.0
-6.0
-4.0
-2.0
0.0
2.0
4.0
6.0
8.0
0 10 20 30 40 50
F1_Transept
F2_Transept
F2_Ambulatory
F2_Trascoro
F2_Trascoro
F3_Transept
F3_Choir
C
8
0
(
d
B
)
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0 10 20 30 40 50
D
5
0
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
0 10 20 30 40 50
G

(
d
B
)
Distance (m)
Fig. 4. Spectral behaviour and average values for acoustic parameters: clarity C
80
,
denition D
50
and sound strength G with respect to source-receiver distance for
receivers with direct sound.
and trascoro during the liturgy on major feasts, as well as the pil-
grims passing through unobtrusively visiting the apostles relics in
the side naves and tribunes, brought about a signicant increase
in sound absorption and a notable reduction in reverberation time,
reaching T
m
values of 3.43 s (Fig. 2).
Although the physical reverberation of the space, T
30
, was vir-
tually independent from the position of the sound source, this
was not the case with its EDT (Fig. 3). During liturgical celebra-
tions in the presbytery, from the main altar (S1), EDT values for
a cathedral full of the faithful were slightly higher than those of
T
30
, which indicated greater reverberation. During chanting from
the choir (S2) reverberation varied greatly depending on receiver
position. Thus, in the choir EDT values were very low, 1.20seconds,
suited to the responsorial psalmody which took place in it, while
the average EDT values in the rest of spaces ranged from 3.00 to
3.35 s. Finally, two different types of behaviour were observed for
liturgical readings to the faithful from the leedoiro. The areas of
the trascoro, side naves, choir and tribunes showed averaged EDT
values ranging from 2.45 to 2.70 s, in contrast with the areas fur-
ther away, such as the transept and ambulatory, with greater EDT
values and therefore greater reverberation. The considerable dis-
tances between source and receiver, combined with the lack of
direct soundandearlyreections were the causes of this increase in
reverberation.
This clear spatial differentiation was complemented by energy
parameters of clarity C
80
and denition D
50
, along with sound
strength G (Fig. 4):

with the sound source (S1) placed on the main altar for mass,
the absence of direct sound and early reections in most areas, in
conjunction with the large source-receiver distances, resulted in
poor acoustic conditions for music, with C
80
values below5dB,
except in areas close to the source: choir, ambulatory and
transept. The same occurred with speech, with very lowD
50
val-
ues, below 0.30. The transept area was occupied by the faithful,
making thempart of the liturgy being celebrated fromthe pres-
bytery and the choir. Therefore this was a space for viewing both
theofces andtherelics, despitepoor acoustic conditions, andthe
importance of speech was not so much in its content but in vocal
variations, within an interpretation of the liturgy transformed
into a theatrical representation;

the position of the sound source in the choir (S2) was quite dif-
ferent fromthe perspective of acoustic conditions. The enclosed
space of the choir providedadequate clarity C
80
andideal D
50
val-
ues for choral singing, allowing for the additionof different voices
and the development of polyphony. The wide stone choir could
house the 72 Chapter members, increasing the number of simul-
taneous soundsources inchoral masses. Thanks tothe amplifying
effects of this closed space, their voices were projected with a
high strength in the cathedral. However, in the remaining areas
of the cathedral this choral singing produced sound with high
reverberationbut little intelligibility (D
50
below0.30), whichwas
partially corrected by the melismatic and syllabic interpretation
of Gregorian chant. The perception of the sound level was lower
(G between 2 and 2dB), producing the sensation of distant, yet
well mixed, sound with no clear directionality;

the position of the choir in the middle of the central nave gen-
erated a space behind it, the trascoro, an area that connected
the faithful to the rite through readings and sermons from the
leedoiro (S3). The high position and the reduction of the source-
receiver distance barely improved speech intelligibility. The use
of cantillation, the chanting of text at a slow rhythm, emphasiz-
ing stress and punctuation, improved intelligibility by adjusting
the sound to the reverberation time, since the conditions in the
trascoro were suited to chant, as can be deduced fromthe values
for C
80
.
The pilgrims walked through the side naves and tribunes see-
ing fragments of the liturgy, and receiving weakened sound from
the different foci of celebration, with little clarity (C
80
outside
the range 5dB), and poor intelligibility (D
50
below 0.10), acous-
tics that were characteristic of a late reverberant eld. However,
pilgrims interacted acoustically with the space, providing sound
absorption, joining in with the liturgical chants of the clergy and
providing responses to the verses of some of the musical pieces,
such as the Dum Paterfamilias, as featured in the Codex. The mass
participation of the pilgrims transformed the Romanesque cathe-
dral of Santiago into a large sound eld.
Thanks to the new technologies used for the simulation of the
acoustic model, auralizationallows us to reproduce what the sound
perceived by a listener in a specic point of the Romanesque cathe-
dral would be, assuming a sound source emitting from another
point of the cathedral. In addition, auralization makes it possible
to carry out a subjective evaluation that can be compared with
objective values. This will be dealt with in future studies. Two
additional audio les are included, assuming the hypotheses of
the cathedral when empty and occupied, for the assessment of the
inuence of the faithful and pilgrims on the acoustic behaviour of
the Romanesque cathedral of Santiago. In both cases the sound
source was located in the choir (S2) and the receiver in R4
(S2 R4 empty.wav and S2 R4 occupied.wav). The audio le is the
result of the convolution of an anechoic recording with binaural
simulated responses, processed with Multivolver software, a tool
supplementing CATT-Acoustic. This anechoic signal consisted of a
fragment of the piece Almighty and Everlasting God, by Orlando
Gibbons [18].
Please cite this article in press as: R. Surez, et al., Intangible cultural heritage: The sound of the Romanesque cathedral of Santiago de
Compostela, Journal of Cultural Heritage (2014), http://dx.doi.org/10.1016/j.culher.2014.05.008
ARTICLE IN PRESS
G Model
CULHER-2864; No. of Pages 5
R. Surez et al. / Journal of Cultural Heritage xxx (2014) xxxxxx 5
5. Conclusions
The layout of the Romanesque cathedral of Santiago de Com-
postela was that of a complex organismwith different interrelated
parts functioning as an enclosed space with two clear functions: as
a church for the faithful and the clergy, with a dual celebration of
the Eucharist in the altar and choir; and the worship of the relics
of the apostles, processing through the side naves and tribunes.
This analysis of the acoustic dimension through the reconstruc-
tion of its sound adds to the knowledge of its intangible heritage
value.
The high reverberation time in the empty church, calculated at
mid frequencies as 8.15 s, was reduced to 3.43 s, thanks to the
sound absorption of the faithful when it was occupied. The lack
of direct sound and nearby early reections in many parts of the
cathedral, added to the presence in many areas of source-receiver
distances over 2025m, causeda major loss inspeechintelligibility
and musical clarity.
The central position of the choir, in architectural and liturgi-
cal terms, had a major repercussion on the acoustic behaviour
of the whole. An ecclesiola in ecclesia was created, a church for
the clergy (actors) within a church for the faithful (spectators),
a double function merging in a single space. With this layout
the sound from the choir space could be heard throughout the
nave, and its multiple reections permeated the remaining spaces.
The faithful identied this central space for liturgical celebra-
tion through the reverberation of sounds with no directionality,
given that it was impossible to see the liturgy and the lack of
direct sound made it hard to locate the source. The reverbera-
tion of sound made the visual experience recognizable, enriching
it.
The space was congured as a liturgical space for music, with
acoustic conditions suited to the modulated Gregorian chant and
the chanted liturgy that was not focused on the intelligibility of the
oral message but on the sensorial perception of reverberation. The
interior reecting space and its great volume expressed the divine
presence through the multi-sensory experience, sound reections,
the mysterious power of being heard without being seen, with a
sound that seemingly came fromeverywhere, blending the voices
of the singers with angelic voices fromheaven, as described in book
IV of the Codex.
Acknowledgements
This work has been nancially supported by the Span-
ish Ministry of Economy and Competitiveness, with reference
BIA2010-20523, and by the Ministry of Education, Culture and
Sport, with the reference FPU12/04949.
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/
j.culher.2014.05.008.
References
[1] E. Chng, Experiential archaeology: is virtual time travel possible? J. Cult. Herit.
10 (2009) 458470.
[2] S. de Blaauw, Architecture and liturgy in Late Antiquity and the Middle Ages.
Traditions and trends in modern scholarship, Archiv fr Liturgiewissenschaft
33 (1991) 134.
[3] P. Navacus Palacio, La catedral de Espa na: Arquitectura y Liturgia, Lunwerg,
Barcelona, 2004.
[4] J. Lpez-Calo, C. Villanueva, El Codex Calixtinus y la msica de su tiempo, in:
Symposium Proceedings, Santiago de Compostela, September 2023, 1999,
Fundacin Pedro Barri de la Maza, A Coru na, 2001.
[5] M. Vecco, A denition of cultural heritage: fromthe tangible to the intangible,
J. Cult. Herit. 11 (2010) 321324.
[6] E. Cirillo, F. Martellotta, Worship, acoustics andarchitecture, Multi-Science U.K,
Brentwood, Essex, 2006.
[7] J. Rindel, C. Christensen, A. Gade, The acoustical history of Hagia Sophia
Re-vivedthroughcomputer simulation, in: Proceedings of the Forum Acous-
ticum, Sevilla September 1620, Sociedad Espa nola de Acstica, Madrid,
2002.
[8] R. Surez, J.J. Sendra, J. Navarro, A.L. Len, The sound of the Cathedral-Mosque
of Crdoba, J. Cult. Herit. 6 (4) (2005) 307312.
[9] J. Navarro, J.J. Sendra, S. Mu noz, The Western Latin church as a place for
music and preaching: an acoustic assessment, Appl. Acoustic 70 (6) (2009)
781789.
[10] D. Howard, L. Moretti, Sound andspace inRenaissance Venice: Architecture,
music, acoustics, Yale University Press, NewHaven, 2009.
[11] R. Surez, J.J. Sendra, A. Alonso, Acoustics, liturgy and architecture in the Early
Christian Church. From the Domus ecclesiae to the Basilica, Acta Acustica
United Acustica 99 (2013) 292301.
[12] B-I.L. Dalenbck, CATT-Acoustic v9 powered by TUCT users manual, CATT,
Gothenburg, Sweden, 2011.
[13] Anonymous, Codex Calixtinus (Englishtranslationavailable at), 2013 (accessed
on October, 2013) https://sites.google.com/site/caminodesantiagoproject/
[14] K.J. Conant, The earlyarchitectural history of the Cathedral of Santiago de Com-
postela, Cambridge/Mass, 1926.
[15] A. Wunderwald, J. Rffer, Die Kathedrale von Santiago de Compostela.
KonstruktionGestaltProgramm, Mitteilungen der Carl Justi-Vereinigung 20
(2008) S.134S.138.
[16] J. Dagenais, Romanesque Redivivus: a full-scale 3D computer reconstruc-
tion of the Medieval Cathedral and town of Santiago de Compostela (ca.
1211), 2013 (accessed in October, 2013 http://etc.ucla.edu/research/projects/
compostela/
[17] ISO-3382-1, Acoustics Measurement of room acoustic parameters Part 1:
performance spaces, 2009.
[18] O. Gibbons, Almighty and everlasting god, sung by St. Olaf Cantorei, 2004. Ane-
choic choral recordings provided by Wenger Corporation. Audio CD and Data
DVD, 2004.